RU2571537C2 - Method to optimise device structure - Google Patents

Abstract

FIELD: measurement equipment.

SUBSTANCE: invention relates to control and measurement equipment and may be used in design of devices for non-destructive testing, assessment and forecasting of condition of objects, structures and engineering facilities, during the entire period of operation. The technical result of the invention, which consists in optimisation of the device structure, is achieved due to the method of optimisation of the device structure. The objective is solved by provision of the device in the form of a graph and searching in it for shortest paths according to the selected optimisation criterion, taking into account energy consumption and/or cost of functional elements.

EFFECT: invention reduces energy consumption of a device and/or cost by reduction of structural redundancy.

7 dwg

Description

The invention relates to measuring equipment and can be used in the design of non-destructive testing devices, assessing and predicting the state of objects, structures and engineering structures throughout the entire period of their operation.

For example, to predict the occurrence of ice on the roads, the Federal Road Agency ordered the following parameters of meteorological conditions and the surface condition of the road surface to be measured [Decree of the Federal Road Agency of November 25, 2009 No. 493-r, “On the publication and application of ODM 218.28.003-2009 ″ Guidelines for specialized forecasting of the condition of the road surface ″]: temperature and relative humidity, surface temperature of the road surface and under it at a depth of 4-7 cm, humidity of the surface of the road surface, etc. The determination of temperature and relative humidity is carried out using automatic road meteorological stations located next to the road. The parameters of the state of the road surface are determined by wireless sensors with self-powered, installed in the upper layer of the roadway.

A known method for remote monitoring and diagnostics of the state of structures and engineering structures and a device for its implementation (RF patent No. 2247958, G01M 5/00, 2003), which consists in the fact that at the control point register signals from measurement units installed in the places of diagnosis of structures, and compare them with pre-fixed values. In this case, the measurement units are mounted on a structural element made of the same material as the entire structure. Metrological certification of the structural element is carried out by establishing dependencies between the signals from the measurement units and the calibrated external influences, register them at the control point and use them as pre-recorded signals. An element with the measurement units mounted on it is cut into the places of diagnosis of the structure, and the state of the structure is judged by the deviation of the received signals from the measurement units from pre-registered signals. The device contains a control point, measurement units located in the places of diagnosis of the structure, converters, communication line, controller. Moreover, the measurement units are placed on a metrologically certified structural element, made of the same material as the entire structure, and embedded in the diagnostic places of the structure. The structural element is connected to the corresponding converters connected by their outputs to the input of the controller connected to the modem, which is connected via the output line to the control point via a communication line.

In the known technical solutions, one of the main elements is the measurement units and transducers, which are used as deformation, linear shear, pressure, vibration, temperature, humidity, flow sensors, etc. In this case, the converters are based on various physical principles. Their main disadvantages are high power consumption and low reliability.

High energy consumption is associated with the presence of power sources (batteries and accumulators) and the operation of sensors and converters in continuous mode. Since the process, for example, of lowering the temperature of the roadway is very slow and only in urgent situations requires a continuous flow of information, the algorithm of the known method and device allows you to set the polling intervals of measurement units from 5 seconds to 1 month. In most cases, 1-2 measurements per day are sufficient to monitor the condition of the road surface. The low reliability of known measurement units and transducers is associated with the reliability of power supplies and the need for their regular replacement. As a rule, the duration of any power source does not exceed several weeks.

A known method for remote monitoring and diagnostics of the state of the structure and engineering structures (RF patent 2471161, G01M 7/00, 2012), which consists in the fact that at the control point, signals are recorded from the measurement units installed in the places of diagnosis of the structure. They are compared with pre-recorded values and judging by the deviation of the received signals from pre-recorded ones about the presence of changes in the controlled parameters. In this case, a structural element is made of the same material as the entire structure. The measurement units are placed on it, metrological certification of the element with the measurement units placed on it is carried out by establishing the relationships between the signals from the measurement units and the calibrated external influences. These dependencies are recorded at the control point and use them as pre-recorded signals. An element is inserted with the measurement units installed on it into the diagnostic locations of the structure and the state of the structure is judged by the deviation of the received signals from the measurement units from pre-registered signals. In this case, the measurement units and transducers are performed in the form of delay lines on surface acoustic waves. On the controller, m harmonic oscillations are formed sequentially at different carrier frequencies, they irradiate delay lines tuned to m carrier frequencies. On each delay line, electromagnetic harmonic vibration is converted into an acoustic wave, ensuring its propagation over the surface of the piezocrystal and back reflection. The reflected acoustic wave is converted into an electromagnetic signal with phase shift keying, the internal structure of which corresponds to the structure of the interdigital transducers of surface acoustic waves, which reflects the serial number of the delay line and the value of the controlled parameter. A complex signal with phase shift keying is radiated into the ether, received at the controller, synchronized detection is performed, a low-frequency voltage is allocated corresponding to the serial number of the delay line and the phase shift corresponding to the external influence, and it is sent to a microprocessor with a storage device in which the received signal is calculated and converted information.

This method, in comparison with other technical solutions of a similar purpose, provides an increase in the effectiveness of remote monitoring and diagnostics of the state of structures and engineering structures, for example, potentially hazardous sections of pipelines, during the entire period of their operation. This is achieved by reducing energy consumption and improving the reliability of sensors, which are one of the main elements of the device that implements the proposed method. The main feature of the sensors in the form of delay lines on surface acoustic waves is their small size and lack of power sources (batteries, accumulators). However, their use significantly complicates the device that implements the indicated method — sequentially connected synchronizer, carrier frequency synthesizer, duplexer, the input-output of which is connected to the transceiver antenna, high-frequency amplifier, and phase detector are added to the device circuit. In addition, the inclusion in the circuit of additional elements and delay lines on surface acoustic waves increases the cost of the device for remote monitoring and diagnostics of the state as a whole.

Given the necessary frequency of measurements and the uniformity of the functional elements of the devices for remote monitoring and diagnostics of the state (Fig. 1), it is possible to reduce the structural redundancy of such devices using the same functional elements in several measuring circuits (measuring circuits of several parameters) by performing their sequential switching. A consequence of a reduction in the structural redundancy of devices is a reduction in their energy consumption and cost.

Thus, as a criterion for optimizing the structure of a device, energy consumption, cost, or a generalized indicator determined by the ideal point method can be used:

$$k_{ij}=\sqrt{{\left(E_{ij}^{n\mathrm{about}Rm}-E_{ij}^{\mathrm{and}d}\right)}^2+{\left(S_{ij}^{n\mathrm{about}Rm}-S_{ij}^{\mathrm{and}d}\right)}^2+{\left(T_{ij}^{n\mathrm{about}Rm}-T_{ij}^{\mathrm{and}d}\right)}^2},\left(\mathrm{one}\right)$$

, $$S_{ij}^{норм}$$

, $$T_{ij}^{норм}$$

- нормированные значения энергопотребления, стоимости и технической совместимости функционального элемента i-го параметра j-го блока соответственно; $$E_{ij}^{ид}$$

, $$S_{ij}^{ид}$$

, $$T_{ij}^{ид}$$

- идеальные значения указанных характеристик ( $$E_{ij}^{ид}$$

=0, $$S_{ij}^{ид}$$

=0, $$T_{ij}^{ид}$$

=1).where i is the measured parameter (i = 1 ... M); M is the maximum number of parameters; j is the block number of the same type of functional elements in the structure of the device (j = 1 ... N); N is the maximum number of blocks of functional elements; $$E_{ij}^{n\mathrm{about}Rm}$$

, $$S_{ij}^{n\mathrm{about}Rm}$$

, $$T_{ij}^{n\mathrm{about}Rm}$$

- normalized values of energy consumption, cost and technical compatibility of the functional element of the i-th parameter of the j-th block, respectively; $$E_{ij}^{\mathrm{and}d}$$

, $$S_{ij}^{\mathrm{and}d}$$

, $$T_{ij}^{\mathrm{and}d}$$

- ideal values of the indicated characteristics ( $$E_{ij}^{\mathrm{and}d}$$

= 0, $$S_{ij}^{\mathrm{and}d}$$

= 0, $$T_{ij}^{\mathrm{and}d}$$

= 1).

The normalization of energy consumption and cost can be carried out as follows:

, $$S_{ij}^{норм}=c_S{S}_{ij}-d_S$$

, $$E_{ij}^{n\mathrm{about}Rm}=c_E{E}_{ij}-d_E$$

, $$S_{ij}^{n\mathrm{about}Rm}=c_S{S}_{ij}-d_S$$

,

и $$c_S=\frac{1}{S_{\max }-S_{\min }}$$

- масштабирующие коэффициенты; $$d_E=\frac{E_{\min }}{E_{\max }-E_{\min }}$$

и $$d_S=\frac{S_{\min }}{S_{\max }-S_{\min }}$$

- коэффициенты сдвига; E_max и S_max - наибольшие энергопотребление и стоимость i-го параметра j-го блока; E_min и S_min - наименьшие энергопотребление и стоимость i-го параметра j-го блока.where E _ij is the power consumption of the i-th parameter of the j-th block; S _ij is the cost of the i-th parameter of the j-th block; $$c_E=\frac{\mathrm{one}}{E_{\max }-E_{\min }}$$

and $$c_S=\frac{\mathrm{one}}{S_{\max }-S_{\min }}$$

- scaling factors; $$d_E=\frac{E_{\min }}{E_{\max }-E_{\min }}$$

and $$d_S=\frac{S_{\min }}{S_{\max }-S_{\min }}$$

- shift factors; E _max and S _max - the largest power consumption and cost of the i-th parameter of the j-th block; E _min and S _min - the lowest power consumption and the cost of the i-th parameter of the j-th block.

The value of technical compatibility is found expertly in the range T _ij ∈ (0; 1] and determines the correspondence of two mating functional elements of the j-th and (j-1) -th blocks according to a number of important technical characteristics, such as accuracy, sensitivity, capacity, complexity installation and others.

Optimization of the device structure according to the selected criterion, for example (1), can be carried out on the basis of graph theory using the shortest path search algorithm, such as Dijkstra or Bellman-Ford algorithm, therefore, the closest in technical essence to the claimed method and selected as a prototype is way to find the shortest path [N. Christofides Graph theory. Algorithmic approach. M .: Mir, 1978.- 432 p. - S.177-183], which consists in the fact that in the graph consisting of V nodes connected by E edges, the source node 1 is determined, it is assigned the label [0; -], where the first element of the label represents the shortest distance to the source node 1, and the second element, the node preceding the shortest path to the source node 1, include it in the set of marked nodes Y = {1}, for each node ν _ij ∉ Y, where i = 1 ... M is the number of the node ν _ij at the jth step; j = 1 ... N is the step number, the sum of the bond weights is established for the path from the given node to the initial D (ν _ij ) = c (1, ν _ij ), for each step the node ν _ij ∉ Y is found for which D (ν _ij ) minimally, assign the label [D (ν _ij ) + c (ν _{i (j-1)} , ν _ij ); ν _{i (j-1)} ] and add the node ν _ij to the set Y, update D (ν _ij ) for all nodes ν _ij ∉ Y according to the rule D (ν _ij ) = min {D (ν _ij ) + c (ν _{i (j-1)} , ν _ij )} until all nodes are in the set Y, determining the shortest routes from node ν _ij for which j = N, to the source node 1 through node passing in the opposite direction using the information provided in the met ah.

The disadvantage of the prototype method is the inability to use it to optimize the structure of devices, due to the following reasons:

1) the prototype method does not provide for the comparison of the graph on which the shortest path is searched with specific functional elements and the relationships between them;

2) the possibility of periodically measuring the parameters and the uniformity of the functional elements of devices mean the inclusion of switching elements in the structural diagram. However, the prototype method does not imply the exclusion of elements that, as a result of determining optimal routes, have one input.

The objective of the invention is to develop a method for optimizing the structure of the device, which allows to reduce its energy consumption and / or cost by reducing structural redundancy.

In the claimed method, this problem is solved by the fact that in the method of optimizing the structure of the device, which consists in the fact that in the graph consisting of V nodes interconnected by E edges, the source node 1 is determined, assign it a label [0; -], where the first the label element is the shortest distance to the source node 1, and the second element is the node preceding the one considered on the shortest path to the source node 1, include it in the set of marked nodes Y = {1}, for each node ν _ij ∉ Y, where i = 1 ... M is the node number ν _ij at the jth step; j = 1 ... N is the step number, the sum of the bond weights is established for the path from the given node to the initial D (ν _ij ) = c (1, ν _ij ), for each step the node ν _ij ∉ Y is found for which D (ν _ij ) minimally, assign the label [D (ν _ij ) + c (ν _{i (j-1)} , ν _ij ), ν _{i (j-1)} ] to it and add the node ν _ij to the set Y, update D (ν _ij ) for all nodes ν _ij ∉ Y according to the rule D (ν _ij ) = min {D) (ν _ij ) + c (ν _{i (j-1)} , ν _ij )} until all nodes are in the set Y , determine the shortest routes from the nodes ν _ij , for which j = N, to the initial node 1 by passing the nodes in the opposite direction using the information presented in met kah, in addition, the graph describes the structure of the device, while assigning the odd nodes of the graph, for which j = 2n + 1, the functional elements of the device, the even nodes of the graph, for which j = 2n, the switching elements, the edges - the connections between functional and switching elements, and vertex weights - a criterion for optimizing the structure of the device. After the shortest routes are determined from the graph, remove all nodes ν _ij for which two conditions are simultaneously satisfied: j = 2n, and the number of edges connecting this node ν _ij with nodes ν _{i (j-1)} is 1, put in correspondence to the nodes of the obtained graph functional and switching elements of the device, edges - the relationship between functional and switching elements.

A new set of essential features allows you to achieve the specified technical result through the use of Dijkstra's algorithm to search for the optimal structure of a digital system according to a given criterion.

The analysis of the prior art made it possible to establish that there are no analogues that are characterized by a set of features identical to all the features of the claimed method for determining the optimal structure of a digital system. Therefore, the claimed invention meets the condition of patentability ″ novelty ″.

Search results for known solutions in this and related fields of technology in order to identify features that match the distinctive features of the claimed object from the prototype showed that they do not follow explicitly from the prior art. The prior art also did not reveal the popularity of the impact provided by the essential features of the claimed invention, the transformations on the achievement of the specified technical result. Therefore, the claimed invention meets the condition of patentability ″ inventive step ″.

The claimed invention is illustrated by the following figures:

figure 1 is a structural diagram of a wireless sensor for measuring several parameters;

figure 2 is a graph of a multifunctional wireless sensor, combined with its structural diagram;

figure 3 - values of technical compatibility of structural elements;

4 is a graph of a wireless sensor with M = 4;

5 is a graph of a multifunctional wireless sensor with optimal routes;

6 is a graph of a multifunctional wireless sensor without switching devices with one input;

7 is an optimal structural diagram of a multifunctional wireless sensor.

The implementation of the claimed method is illustrated by the example of designing a wireless sensor with autonomous power for measuring several parameters (figure 1).

Given the monotony (slow change over time) of the parameters of the state of the road surface, it is possible to periodically measure them and it is proposed to use the same functional elements in the structural diagrams of several wireless sensors to reduce energy consumption by performing their serial switching. To do this, perform the following steps.

1. The structure of the device is described by a graph, and the odd nodes of the graph, for which j = 2n + 1, are the functional elements of the device, the even nodes of the graph, for which j = 2n, are the switching elements, and the edges are the connections between the functional and switching elements (figure 2), and the weights of the vertices - the criterion for optimizing the structure of the device.

In the considered example (Fig. 2), as a criterion for optimizing the structure of the device, we choose the generalized indicator (1), then c (ν _{i (j-1)} , ν _ij ) = k _ij . The technical compatibility values of the interacting structural elements of the multifunctional wireless sensor for the case of measuring four parameters (M = 4) are presented in figure 3, and the corresponding graph is shown in figure 4.

2. In the graph consisting of V nodes interconnected by E edges, the source node 1 is determined, the label [0; -] is assigned to it, where the first label element is the shortest distance to the source node 1, and the second element is the node preceding considered on the shortest path to the source node 1, include it in the set of marked nodes Y = {1}, for each node ν _ij ∉ Y, where i = i ... M is the number of the node ν _ij at the jth step; j = 1 ... N is the step number, the sum of the bond weights is established for the path from the given node to the initial D (ν _ij ) = c (1, ν _ij ), for each step the node ν _ij ∉ Y is found for which D (ν _ij ) minimally, assign the label [D (ν _ij ) + c (ν _{i (j-1)} ), ν _ij ); ν _{i (j-1)} ] and add the node ν _ij to the set Y, update D (ν _ij ) for all nodes ν _ij ∉ Y according to the rule D (ν _ij ) = min {D (ν _ij ) + c (ν _{i (j-1)} , ν _ij )} until all nodes are in the set Y , determine the shortest routes from nodes ν _ij , for which j = N, to the source node 1 by passing the nodes in the opposite direction using the information presented in met kah.

The results of the iterative process considered are presented in Fig. 5, where the nodes show the sum of the bond weights for the path from this node to the original D (ν _ij ) = c (1, ν _ij ), and the optimal routes are from all the initial nodes (i = 1 ) to the original node 1 are highlighted in bold. The obtained routes can eliminate the structural redundancy of the multifunctional wireless sensor (figure 2), removing 20 functional elements.

An analysis of the graph (Fig. 5) shows that between the functional elements of the wireless sensor switching elements (nodes with j = 2n, n = 1,2 ...) having only one input are included. Their use is impractical and requires exclusion from the graph (structural diagram of a wireless sensor).

3. From the obtained graph, remove all nodes ν _ij for which two conditions are simultaneously satisfied: j = 2n, and the number of edges connecting this node ν _ij with nodes ν _{i (j + 1)} is 1.

So, for the case under consideration (Fig. 5), nodes ν ₁₂ , ν ₃₂ , ν ₄₂ , ν ₁₃ , ν ₃₃ , ν ₁₄ , ν ₂₄ , ν ₃₄ , ν ₄₄ , ν ₁₅ , ν ₄₅ , ν ₁₆ are excluded at this step , ν ₃₆ , ν ₄₆ , ν ₁₇ , ν ₃₇ , ν ₄₇ , ν ₁₈ , ν ₂₈ , ν ₃₈ , ν ₄₈ , ν ₂₉ , ν ₃₉ , ν ₄₉ , ν ₁₁₀ (FIG. 6).

4. The functional and switching elements of the device are aligned with the nodes of the obtained graph, and the edges between the functional and switching elements are connected to the edges, resulting in a combination of functional and switching elements of a device with an optimized structure.

For the considered example, taking into account specific functional and switching elements, the device structure (wireless sensor) that is optimal according to the selected criterion (1) can be represented as follows (Fig. 7).

The gain in the generalized indicator from the implementation of the developed scheme (Fig. 7) compared with the initial one (Fig. 1) was about 2.2 times, which for

for the values of the generalized indicator k _ij specified in figure 3, and

for existing values of k _ij given in FIG. 6, corresponds to the calculated value

Thus, the proposed method for optimizing the structure of the device allowed (in the considered example) to reduce the structural redundancy of the wireless sensor by 35% and achieve a gain in the generalized indicator of 2.35 times. The gain value, taking into account the real energy consumption of the functional and switching elements of the implemented device (wireless sensor), was about 2.2 times.

Claims (1)

A way to optimize the structure of the device, which consists in the fact that in the graph consisting of V nodes interconnected by E edges, the source node 1 is determined, it is assigned the label [0; -], where the first label element represents the shortest distance to the source node 1 , and the second element is the node preceding the shortest path to the source node 1, include it in the set of marked nodes Y = {1}, for each node ν _ij ∉ Y, where i = 1 ... M is the number of the node ν _ij on j m step; j = 1 ... N is the step number, the sum of the bond weights is established for the path from the given node to the initial D (ν _ij ) = c (1, ν _ij ), for each step the node ν _ij ∉ Y is found for which D (ν _ij ) minimally, assign the label D (ν _ij ) + c (ν _{i (j-1)} , ν _ij ); ν _{i (j-1)} ] and add the node ν _ij to the set Y, update D (ν _ij ) for of all nodes ν _ij ∉ Y by the rule

until all nodes are in the set Y, the shortest routes are determined from the nodes ν _ij , for which j = N, to the initial node 1 by passing the nodes in the opposite direction using the information presented in the labels, characterized in that the graph describe the structure of the digital system, while assigning the odd nodes of the graph, for which j = 2n + 1, the functional elements of the digital system, the even nodes of the graph, for which j = 2n, the switching devices, the edges - the connections between the functional elements and the switching devices and, as the balance of vertices - optimization criterion after determining the shortest routes from the obtained graph removed all nodes ν _ij, for which two conditions are simultaneously satisfied: j = 2n, and the number of edges connecting this node ν _ij with ν _i nodes _{( j + 1)} , equal to 1, correspond to the nodes of the obtained graph the functional elements of the digital system and switching devices, and the edges - the relationship between the functional elements and switching devices.

Images